U.S. patent application number 09/469122 was filed with the patent office on 2003-03-27 for amorphous silicon sensor with micro-spring interconnects for achieving high uniformity in integrated light-emitting sources.
Invention is credited to CHUA, CHRISTOPHER L., FORK, DAVID K., LEMMI, FRANCESCO, LU, JENGPING, MCINTYRE, HARRY J., MEI, PING.
Application Number | 20030057533 09/469122 |
Document ID | / |
Family ID | 23862508 |
Filed Date | 2003-03-27 |
United States Patent
Application |
20030057533 |
Kind Code |
A1 |
LEMMI, FRANCESCO ; et
al. |
March 27, 2003 |
AMORPHOUS SILICON SENSOR WITH MICRO-SPRING INTERCONNECTS FOR
ACHIEVING HIGH UNIFORMITY IN INTEGRATED LIGHT-EMITTING SOURCES
Abstract
A hybrid structure or device is provided wherein carried on a
single substrate is at least one micro-spring interconnect having
an elastic material that is initially fixed to a surface of the
substrate, an anchor portion which is fixed to the substrate
surface and a free portion. The spring contact is self-assembling
in that as the free portion is released it moves out of the plane
of the substrate. Also integrated on the substrate is a sensor
having an active layer and contacts. The substrate and sensor may
be formed of materials which are somewhat partially transparent to
light at certain infrared wavelengths. The integrated sensor/spring
contact configuration may be used in an imaging system to sense
output from a light source which is used for image formation. The
light source may be a laser array, LED array or other appropriate
light source. The sensor is appropriately sized to sense all or
some part of light from the light source. The sensor may also be
sufficiently transparent so that light is not blocked from its
emission path, with a contrast ratio such that it only absorbs a
small fraction of light passing therethrough. An additional
characteristic is that the manufacturing process is compatible with
the manufacturing process for the micro-spring interconnects. Data
from the sensor is used as light source correction information.
This information is provided to a calibration configuration which
allows for calibration of high-speed systems.
Inventors: |
LEMMI, FRANCESCO; (MENLO
PARK, CA) ; CHUA, CHRISTOPHER L.; (SAN JOSE, CA)
; MEI, PING; (PALO ALTO, CA) ; LU, JENGPING;
(MOUNTAIN VIEW, CA) ; FORK, DAVID K.; (PALO ALTO,
CA) ; MCINTYRE, HARRY J.; (OCEANSIDE, CA) |
Correspondence
Address: |
MARK S. SVAT
FAY SHARPE FAGAN MINNICH & MCKEE LLP
1100 SUPERIOR AVENUE
SEVENTH FLOOR
CLEVELAND
OH
441142518
|
Family ID: |
23862508 |
Appl. No.: |
09/469122 |
Filed: |
December 21, 1999 |
Current U.S.
Class: |
257/678 ;
257/E25.02; 257/E31.062 |
Current CPC
Class: |
H01L 2924/00 20130101;
H01L 2924/0002 20130101; H01L 25/0753 20130101; H01S 5/02 20130101;
H01S 5/0262 20130101; H01L 2924/3011 20130101; H01S 5/4025
20130101; H01L 31/1055 20130101; H01L 31/1016 20130101; B41J 2/45
20130101; H01L 2924/0002 20130101 |
Class at
Publication: |
257/678 |
International
Class: |
H01L 023/02 |
Claims
Having thus described the invention, it is now claimed:
1. A hybrid device comprising: a substrate; a micro-spring
interconnect formed on the substrate, the micro-spring interconnect
including, an elastic material that is initially fixed to a surface
on the substrate including, an anchor portion fixed to the
substrate, and a free portion; and a sensor formed on the
substrate, the sensor including an active layer and contacts, said
active layer being capable of sensing light, said micro-spring
interconnect and said sensor being integrated on the substrate.
2. The invention according to claim 1 wherein the hybrid device
further includes at least one of a single light source, an array of
lasers, and an array of light emitting diodes (LEDs), positioned to
emit light at least partially through a portion of the sensor.
3. The invention according to claim 2 wherein the sensor is
designed and aligned with at least one of the laser array and the
LED array, to receive emitted light from at least one of, some of
the lasers of the laser array and some of the LEDs of the LED
array.
4. The invention according to claim 2 wherein the sensor is
designed and aligned with at least one of the laser array and the
LED array to receive emitted light from a portion of at least one
of the laser array and the LED array.
5. The invention according to claim 4 wherein the substrate is
designed and aligned with at least one of the laser array and the
LED array to receive emitted light from a portion of at least one
of the laser array and the LED array.
6. The invention according to claim 1 wherein the sensor is an
array of sensors.
7. The invention according to claim 1 wherein the substrate and the
active layer of the sensor at least partially transparent at
selective wavelengths.
8. The invention according to claim 1 wherein the sensor and the
micro-spring interconnect are comprised of materials which allow
for integration of the micro-spring interconnect and the sensor on
the single substrate during a manufacturing process, wherein at
least one of the materials for the micro-spring interconnect and
the sensor is the same.
9. The invention according to claim 1 wherein the sensor is
comprised of, a first transparent/conductive layer; an active layer
on top of the first transparent/conductive layer; a second
transparent/conductive layer on top of the active layer; a
passivation/release layer located over at least the first
transparent/conductive layer and the second transparent/conductive
layer; vias through the passivation/release layer to the first and
second transparent/conductive layers; and a metal layer connecting
to the first and second transparent/conductive layers through the
vias, wherein the metal layer acts as signal lines to receive and
carry signals from the active layer.
10. The invention according to claim 9 wherein the elastic material
fixed to the substrate is held by the passivation/release layer,
which is interposed between the substrate and the elastic
material.
11. The invention according to claim 10 wherein the elastic
material is a stressed metal layer having sub-layers of differing
stress gradients, whereby when released from the
passivation/release layer, the released portion moves out of a
plane of the substrate.
12. The invention according to claim 1 wherein the sensor further
includes an absorption layer, located immediately over the sensor,
wherein the absorption layer absorbs unwanted light prior to being
detected by the active layer.
13. The invention according to claim 9, wherein the active layer is
a three layer element, wherein a first layer is a n+doped amorphous
silicon, the first layer being one of, but not limited to
n+phosphorous-doped amorphous silicon and n+arsenic-doped silicon;
wherein a second layer is an intrinsic amorphous silicon; wherein a
third layer is a p+doped amorphous silicon, the third layer being,
but not limited to, p+boron-doped amorphous silicon.
14. The invention according to claim 1 wherein a switch is located,
between the sensor and the substrate, such that the sensor is an
active semi-continuous sensor.
15. The invention according to claim 14 wherein the switch is a
thin-film-transistor (TFT).
16. The invention according to claim 1 wherein the micro-spring
interconnect is a plurality of micro-spring interconnects.
17. A hybrid device comprising: at least one of a laser or LED
device capable of emitting light at a certain wavelength; a
substrate; a micro-spring interconnect formed on the substrate the
micro-spring interconnect including, an elastic material that is
initially fixed to the substrate, an anchor portion fixed to the
substrate, and a free portion; and a sensor formed on the
substrate, in an integrated manner, with the micro-spring
interconnect, the sensor including an active layer and contacts,
wherein said substrate, and said sensor are at least partially
transparent to light at the wavelength emitted by at least one of
the laser or the LED device; and said at least one of the laser or
the LED device and said substrate with said sensor and said at
least one micro-spring interconnect being separately fabricated and
aligned, such that at least a portion of the light emitted by the
at least one of the laser and LED device is directed through at
least a portion of the substrate and the sensor.
18. The invention according to claim 17, wherein at least a portion
of the laser or the LED device is a plurality of lasers or LEDs
formed in a laser or LED array.
19. The invention according to claim 17 wherein the sensor is sized
such that each of the lasers or LEDs emit light at least partially
through the sensor.
20. The invention according to claim 17 wherein the sensor is a
plurality of sensors, sized such that a sub-group of the lasers or
the LEDs may emit light through selected ones of the of
sensors.
21. The invention according to claim 19 wherein the lasers or LEDs
are arranged as a printbar, and the micro-spring interconnect is in
electrical contact with the printbar.
22. A calibration/printing system comprising: a sensor
configuration including a sensor element integrated on a substrate
with a plurality of micro-spring interconnects; a light source
aligned with the sensor configuration such that at least a portion
of the light from the light source is sensed by the sensor and at
least a first of the micro-spring interconnects is in physical
contact with a portion of the light source; a driver chip aligned
with the sensor configuration and the light source such that at
least a second of the micro-spring interconnects is in physical
contact with a portion of the driver chip, whereby a communication
path is formed between the light source and the driver chip by the
at least first and second micro-spring interconnects.
23. The invention according to claim 22 wherein the driver chip
further includes: a comparator for comparing a sensor readout
current from the sensor and a reference current; a converter
arrangement which converts the output of the comparator into
digital data representing characteristics of the light source; a
set of low frequency shift registers configured to receive and
store the digital data; an activation signal selectively supplied
to the light source to selectively emit light therefrom; a driver
designed to interpret the digital data as activation signal
correction information for the activation signal; a high frequency
shift-register configured to receive and store digital image data
from a source external to the driver chip; and an enable/disable
output from the high frequency shift-register to selectively supply
the activation signal and light source correction information to
the light source, whereby an amount of light emitted by the light
source is controlled.
24. The invention according to claim 22 wherein the digital image
data from the source external to the driver chip is supplied as
high frequency bit stream data.
25. The invention according to claim 22 wherein the light source is
a printbar having an array of light sources, and wherein the
printbar is controlled to activate the light sources in a
sequential manner to obtain calibration data to be stored in the
driver.
26. A hybrid device comprising: a micro-spring interconnect
structure; and at least two devices electrically connected by the
interconnect structure wherein, one of the devices is a sensor, the
sensor including an active layer and contacts, said active layer
being capable of sensing light, and another one of the devices is
at least one of a single light source, an array of lasers, and an
array of light emitting diodes (LEDs), positioned to emit light at
least partially through the sensor.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] This invention relates to imaging systems using a sensor
element to control the emission of at least some of the light of
one or more light-emitting sources. In particular, this invention
is directed to architectures, characteristics and methods of
integration of a light sensor configuration with different light
sources, using micro-fabricated metal spring contacts.
[0003] 2. Technical Background
[0004] Image printbars which are used in imaging systems are well
known in the art. Such printbars are generally comprised of a
linear array of a plurality of discreet, light-emitting sources.
Examples of such printbars include light-emitting diodes and
lasers. A method of forming lasers in semiconductor material, which
may be used in the formation of laser printbars has been taught by,
for example, U.S. Pat. No. 5,978,408 to Thornton, entitled, "Highly
Compact Vertical Cavity Surface Emitting Lasers", issued Nov. 2,
1999; and U.S. Pat. No. 5,843,802 to Beernink, et al., entitled,
"Semiconductor Laser Formed by Layer Intermixing", issued Dec. 1,
1998, both commonly assigned and hereby incorporated by
reference.
[0005] In a typical printbar arrangement, a large number of
individual light-emitting sources are arranged in an elongated,
planer array that is placed adjacent an image recording member. By
providing relative motion between the printbar and the image
recording member, the printbar scans the image recording member,
and by selectively illuminating the individual light-emitting
sources, a desired light image is recorded on the image recording
member.
[0006] The selective illumination of the individual light-emitting
sources is performed according to image-defining data that is
applied to printbar driver circuitry. Conventionally, the
image-defining data takes the form of simple binary video data
signals. Those data signals may be from any of a number of data
sources such as a Raster Input Scanner (RIS), a computer, a word
processor, or a facsimile machine. Typically, the binary video data
is clocked into a shift register. After completely shifting the
data into the shift register, the contents of the shift register is
transferred in parallel into latch circuits for temporary storage.
Then, upon the occurrence of a start of a line signal, the latch
data is applied to the printbar drive circuit which thereby
illuminates the individual light-emitting sources of the printbar
so as to produce a line of the latent image. A complete latent
image is formed by performing successive line exposures until the
image is produced. Due to their narrow beam profile and high
efficiency, photolithographically configured laser printbars have
been found to provide certain advantages. Proposed laser printbars
consist of an array of Vertical-Cavity Surface-Emitting Lasers
(VCSELs) which may be designed with as small as 3 .mu.m pitch. At
such a pitch, a 4 cm-long laser chip would accommodate more than
13,300 individually addressable laser elements, more than necessary
for 1,200 dpi printing on a standard 11 inch-long paper, where
13,200 elements are required. A drawback of such a large number or
light sources, ultra-high density-packed, is the expectation of
non-uniformity of laser responses. This non-uniformity has the
potential for high spatial frequency that makes the effect on
printed images noticeable to the human eye.
[0007] One manner of addressing non-uniformity is to perform a
calibration when the printbar is being manufactured. A problem with
this process is that it does not address aging of the lasers,
fluctuations in driver chip operation or environmental variations
such as temperature and humidity, among others.
[0008] A second proposal is to form a sensor or detector as part of
the printbar in order to perform periodic calibrations during the
lifetime of the printbar. This concept is described in U.S. Pat.
Ser. No. 08/921,942, entitled, Semiconductor Laser With Integrated
Detector Structure, Thornton et al., filed Aug. 27, 1997. A
drawback with this proposal is the complexity of forming the
device.
[0009] Similar issues may be present in many other imaging systems
where one or more light-emitting sources need to be controlled in
order to address issues like intrinsic non-uniformity, drift of
characteristics or differential aging.
[0010] Therefore, it has been considered desirable to provide an
apparatus and method to integrate a sensor element in a hybrid
structure with a printbar or another compatible light-emitting
source using simple patterned micro-spring metal contacts.
SUMMARY OF THE INVENTION
[0011] Provided is a hybrid structure or device integrated in a
substrate, where in some cases the substrate is substantially
transparent to light at infrared wavelengths. Integrated on the
substrate are a plurality of micro-spring interconnects, where the
micro-spring interconnects are formed of an elastic material that
is initially fixed to a surface on the substrate. Upon release of a
sacrificial layer a free portion moves out of the plane of the
substrate in a self-assembling. A sensor is formed on the same
substrate, and includes an active layer and contacts. The active
layer may be substantially transparent to light at infrared
wavelengths. The micro-spring interconnects and the sensor are
integrated on the substrate and configured using a compatible
manufacturing process.
[0012] With attention to a further embodiment of the present
invention, a light-emitting source is provided which may be an
array of individual light-emitting sources. The light sources may
be lasers such as, Vertical Cavity, Surface-Emitting Lasers
(VCSELs), which are formed on the substrate, and the VCSELs are
capable of emitting light at an infrared wavelength. Other light
sources may also be used such as an array of light emitting diodes
(LEDs). The substrate holding the spring contacts and sensor, and
the substrate including the light sources are aligned such that at
least a portion of the light emitted by the light source is
directed through the second substrate and the sensor which may be,
substantially transparent at infrared wavelengths.
[0013] Separate embodiments describe similar integration schemes
for less directional light-sources, such as Light-Emitting Diodes
(LEDs). It is to be appreciated that the light of other wavelengths
may be used in conjunction with the concepts of this invention.
BRIEF DESCRIPTION OF THE DRAWINGS
[0014] FIG. 1 depicts a portion of a laser lightbar used in
association with the present invention;
[0015] FIG. 2 shows alignment of two components that allow laser
light to pass through a second substrate, partially transparent at
the wavelength of the light;
[0016] FIG. 3 depicts a block diagram of an arrangement according
to the present invention;
[0017] FIG. 4 is a bottom view of FIG. 3;
[0018] FIGS. 5a-5e illustrate process steps for the formation of a
hybrid device according to the teachings of the present
invention;
[0019] FIGS. 6a-6e are top views of FIGS. 5a-5e;
[0020] FIGS. 7a-7b shows selected process steps of a second
embodiment according to the teachings of the present invention;
[0021] FIGS. 8a and 8b are top views of the process steps of FIGS.
7a-7b;
[0022] FIG. 9 depicts a third embodiment in accordance with the
teachings of the present invention;
[0023] FIG. 10 is a block diagram of an arrangement implementing an
LED printbar according to the teachings of the present
invention;
[0024] FIG. 11 is a bottom view of FIG. 10;
[0025] FIG. 12 depicts a block diagram of a further embodiment
implementing an LED printbar according to the teachings of the
present invention;
[0026] FIG. 13 is a bottom view of FIG. 12;
[0027] FIG. 14 illustrates a block diagram of an imaging system
designed for calibration of light signals implementing concepts of
the present invention;
[0028] FIG. 15 depicts timing of light and current readouts used in
the calibration operation of the present invention; and
[0029] FIG. 16 is a schematic of a driver element of the driver
device of FIG. 14.
DETAILED DESCRIPTION OF THE EMBODIMENTS
[0030] The following description will primarily focus on a system
employing a laser printbar. However, it is to be appreciated the
present invention may also be used in conjunction with a LED array
or other appropriate light emitting device or system, also for
purposes other than printing. Further, the following discussion
emphasizes that the described configuration is sufficiently
transparent to be used in sensing output from a printbar. It is to
be appreciated however, the concepts of the present invention may
be used in applications where it is appropriate to have a sensor
which absorbs a higher percentage, and possibly all light emitted
by a light emitting device or system.
[0031] Turning to FIG. 1, illustrated is a section of a laser
printbar 10 having individual lasers 12 interleaved at a 3 .mu.m
pitch spacing.
[0032] Printbar 10 may be designed using gallium arsenide (GaAs),
and lasers 12, in one embodiment may be Vertical-Cavity
Surface-Emitting Lasers (VCSELs). A laser emission cone 14 is shown
for one laser to illustrate that the typical divergence of a VCSEL
beam can be smaller than 20.degree..
[0033] FIG. 2, is a cross-section of a spring contact device 16. A
first portion of spring contact device 16 is silicon or glass
substrate 18 which has patterned thereon micro-spring interconnects
(also called spring contacts) 20 and 22. Device 16, in one
embodiment, further includes printbar 10, having an array of lasers
12, a first driver chip 24 and a second driver chip 26. Each of
driver chips 24 and 26 may control operation of the lines of one
side of the array of lasers 12. Spring contacts 20 and 22 are
designed to provide an electrical connection between driver chips
24, 26 and printbar 10. The electrical connection between chips 24,
26 and printbar 10 can be obtained by bonding these elements to
spring contacts 20 and 22. It is to be noted that although not
shown, printbar 10 and chips 24, 26 may but do not need to be
carried on a further-substrate. Driver chips 24, 26 receive image
data which are converted into signals delivered to printbar 10. The
signal driver chips 24, 26 selectively control operation of lasers
12, such as VCSEL-type lasers, which generate a light beam 28 in
accordance with received image data, and emit the beam 28 through
substrate 18. Therefore, it is necessary that substrate 18 be
partially transparent to light in the frequency range emitted by
lasers 12. In this embodiment, lasers 12 generate a wavelength
shorter than 870 nanometers, in the infrared (IR) range.
[0034] One arrangement of a printbar and spring contacts is
disclosed in U.S. Pat. No. 5,944,537, to Smith et al., entitled,
Photolithographically Patterned Spring Contact and Apparatus and
Methods for Electrically Contacting Devices, issued Aug. 31, 1999,
hereby incorporated by reference.
[0035] A method of packaging devices being contacted with
micro-springs is disclosed in Xerox Patent Application D/99734, to
Chua et al., entitled Method and Apparatus for Interconnecting
Devices Using an Adhesive, filed Dec. 15, 1999, commonly assigned
and hereby incorporated by reference.
[0036] Spring contacts 20 and 22 are photolithographically
patterned on substrate 18 and designed for electrical connections
between devices. An inherent stress gradient in each spring contact
causes free portions of the spring contacts to bend up and away
from the substrate when a sacrificial layer is selectively removed.
An anchor portion remains fixed to the substrate. The spring
contact is made of an elastic material and the free portions
provide for compliant contacts between devices for an electrical
interconnection.
[0037] In one embodiment such contacts are designed in accordance
with the teachings of U.S. Pat. No. 5,613,861 to Smith et al.,
entitled, "Photolithographically Patterned Spring Contact"; U.S.
Pat. No. 5,848,685 to Smith et al., entitled,
"Photolithographically Patterned Spring Contact"; U.S. Pat. No.
5,914,218 to Smith et al., entitled, "Method for Forming a Spring
Contact"; and U.S. Pat. No. 5,944,537 to Smith et al., entitled,
"Photolithographically Patterned Spring Contact and Apparatus and
Methods for Electrically Contacting Devices", all commonly assigned
and hereby incorporated by reference.
[0038] Implementing spring contacts 20, 22, allows printbar 10 to
be bilaterally electrically connected to driver chips 24, 26. When
printbar 10 and driver chips 24,26 are moved to contact under the
construction of FIG. 2, a gap of approximately 20 .mu.m gap
separates the surfaces of elements 10, 24 and 26 from the surface
of the spring contacts' substrate, element 18. For a laser printbar
and arrangement such as described in FIGS. 1 and 2, the issue of
non-uniformity between the many different lasers 12 is a
significant problem.
[0039] The present application describes devices and systems to
correct the non-uniformity in light output, by provision of a
system that uses a sensor that monitors the output of each light
source to assist in calibration operations where the sensor is
integrated on a substrate with spring contacts. The values
resulting from the calibration are stored in an electronic look-up
table, or by some other data storage method that can be referenced
to normalize the output of individual light sources in-situ, by
implementing periodic calibration operations.
[0040] FIG. 3 illustrates an embodiment of the present invention
shown as a hybrid structure 30 similar to what was depicted in FIG.
2, however, in this design a sensor 34 and spring contacts 20, 22
are integrated on the same substrate 32. It is to be noted that
substrate 32 as well as sensor 34 are partially transparent and
laser beam (IR radiation) 28 is capable of passing substantially
unobstructed through substrate 32, and sensor 34. By forming sensor
34 in a fashion which allows it to be aligned with a high degree of
precision in front of lasers 12, it is possible to obtain in-situ
information as to laser output for each of lasers 12, and provide
periodic calibration of printbar 10 for an operational imaging
device.
[0041] FIG. 4 depicts a bottom view of FIG. 3. Sensor 34 and spring
contacts 20, 22 are on substantially the same plane nearest the
page surface, and lasers 12 of printbar 10 for connection to spring
contacts 20, 22 are at the back of the page. Contact pads 36, shown
in FIG. 4, are pre-patterned on printbar 10 for connection with
micro-springs 20, 22. Sensor feedback lines 38, 39 are shown
extending from sensor 34. Sensor feedback lines 38, 39 carry
readout current used in the calibration operation. FIG. 4
emphasizes the importance of alignment between sensor 34 and the
array of lasers 12, and that sensor 34 is sufficiently sized to
cover all lasers 12 in this embodiment.
[0042] FIGS. 5a-5e and 6a-6e are cross-sectional and top views of
the process to form a device having contact springs 20, 22 and
sensor 34 integrated on a single substrate 30. In order to
accomplish this integration, it is necessary for the sensor to be
configured to have certain unique properties. In this embodiment,
the first characteristic is that the sensor be made large enough to
be aligned with all lasers 12 of printbar 10. Presently a printbar
having a laser array of approximately 4 cm by 200 .mu.m is
anticipated for use in an imaging device.
[0043] The second characteristic requires the sensor to be
partially transparent to the laser light. As previously noted, this
requirement allows for the operation of calibration without moving
the printbar out of the printing area.
[0044] The third characteristic is for the sensor to have a high
"contrast ratio", also called "light-to-dark" response. Since the
sensor will only absorb a fraction of the light passing
therethrough, due to its partially transparent nature, it must be
able to work even with very small signals. An ideal sensor will
have no current flowing when no light exists. Amorphous silicon
(a-Si:H) sensors are able to approach this ideal state.
[0045] The fourth characteristic is that the manufacturing process
for the sensor must be compatible with the manufacturing process of
spring contacts.
[0046] Further, in this embodiment, the fabrication process
depicted in FIGS. 5a-5e and 6a-6e, must ensure that when the
integrated device is formed and contacted to a printbar, the sensor
and spring contacts are properly aligned in relation to the lasers
and the driver chips.
[0047] With particular reference to the process flow for the
construction of a device formed on a substrate integrating both
spring contacts and a sensor, in a first stage (stage 1), on a
transparent silicon or glass substrate 40 is deposited or grown a
transparent/conductive layer 42, such as iridium tin oxide (ITO),
which is patterned in accordance with known techniques.
Transparent/conductive layer 42 needs to be transparent such that
it does not block light emitted from lasers 12 (of FIGS. 3 and 4),
and is required to be conductive as it will act as a first
electrode of the sensor. FIG. 6a depicts a top view of stage 1.
[0048] Turning to stage 2, illustrated by FIGS. 5b and 6b, a
hydrogenated amorphous silicon sensor (a-Si:H) component or active
layer 44 is grown on top of the first transparent/conductive layer
42. a-Si:H sensor component 44 is usually comprised of three
layers. The first layer 44a, is a n.sup.+-doped layer of material,
typically less than 1,000 angstroms in thickness. Though not
limited thereto, the first layer 44a may be a n.sup.+
phosphorous-doped amorphous silicon, or n.sup.+ arsenic-doped
silicon. A second layer 44b is intrinsic amorphous silicon, of a
thickness less than a micron, preferably in the range of
3,000-5,000 angstroms. The third layer 44c of sensor element 44 is
a p.sup.+ doped amorphous silicon of approximately 100 angstroms
thickness. An example of a p.sup.+-doped amorphous silicon which
may be used as third layer 44c is p.sup.+ boron-doped amorphous
silicon.
[0049] Following deposition of sensor element 44, a second
transparent/conductive layer 46 is deposited on top of sensor
element 44. Sensor element 44 and second transparent/conductive
layer 46 may be patterned together in a single process or
separately. Further, sensor element 44 in the present embodiment is
an amorphous silicon sensor, which is opaque in visible light, and
transparent at IR wavelengths.
[0050] Turning to step 3 of the process, shown in FIGS. 5c and 6c,
passivation/release layer 48 is deposited. Passivation/release
layer 48 is required to meet the manufacturing requirements of both
the sensor configuration and spring contacts. One type of substance
that meets this requirement of compatibility is amorphous
silicon-nitride. Oxynitride, and polyamide are among other possible
choices. In particular, layer 48 acts as a passivation layer for
the sensor by being electrically insulating and is transparent in
the wavelength range emitted by the lasers which the sensor is to
be associated. Layer 48 also functions as a release/sacrificial
layer that may be used in the configuration of the spring contacts,
as will be described in greater detail below.
[0051] Two vias are provided through passivation/release layer 48
to allow contact to transparent/conductive layers 42 and 46. First
via 50 and second via 52 may be seen clearly in top view FIG. 6c.
The first via 50 provides an opening to second
transparent/conductive layer 46 and second via 52 provides an
opening to first transparent/conductive layer 42. These openings
are necessary since the passivation/release layer 48 is formed from
an electrically insulating material and, since layers 42 and 46 act
as electrodes of the sensor, these openings provide access to
layers 42, 46.
[0052] At this point, an electrically protected sensor 53 is formed
consisting of first transparent/conductive layer 42, sensor element
44, second transparent/conductive layer 46 and passivation/release
layer 48, and vias 50, 52 which provide electrical access to sensor
53.
[0053] Attention is now directed to stage 4 of the process, as
illustrated in FIGS. 5d and 6d. In this stage, metal patterns
54a-n, 56, and 58 are deposited directly onto passivation/release
layer 48 and into vias 50 and 52. Metal patterns 54a-n, 56, and 58,
are deposited during the same processing steps and are made of the
same metal layers, formed to have an inherent stress gradient.
[0054] In one preferred embodiment, metal patterns 54a-n, 56, and
58 are made of an extremely elastic material, such as a
chrome-molybdenum alloy or a nickel-zirconium alloy. Depositing of
the metal patterns 54a-n, 56, and 58 may be achieved by many
methods including electron-beam deposition, thermal evaporation,
chemical vapor deposition, sputter deposition or other methods.
[0055] The metal layers that compose patterns 54a-n, 56 and 58 may
be thought of as deposited in several sub-layers to a final
thickness. A stress gradient is introduced into the metal layers by
altering the stress inherent in each of the sub-layers. Different
stress levels can be introduced into each sub-layer of the
deposited metal during the deposition processing. After metal
layers for patterns 54a-n, 56 and 58 have been deposited, they are
patterned by photolithography into desired designs.
[0056] The process of depositing metal layers for patterns 54a-n,
56 and 58 in separate sub-layers results in the patterns 54a-n, 56
and 58 having a stress gradient which is compressive in a lower
metal layer becoming increasingly tensile toward the top metal
layer. Although the stress gradient urges the metal layers that
compose patterns 54a-n, 56 and 58 to bend into an arc, patterns
54a-n, 56 and 58 adhere to the passivation/release layer 48 and
thus lays flat.
[0057] In step 5, as depicted in FIGS. 5e and 6e, free portions
60a-n of metal patterns 54a-n are released from passivation/release
layer 48 by a process of undercut etching. Passivation/release
layer 48 is typically deposited by plasma chemical vapor deposition
(PECVD) and can give passivation/release layer 48 a fast etch rate
characteristic. After proper photolithography a selective etchant,
typically a HF-based solution, may be used to etch the
passivation/release layer 48. The etchant is called a selective
etchant because it etches passivation/release layer 48 faster than
the selective etchant removes metal from metal patterns 54a-n. By
means of the etch process free portions 60a-n are released from
passivation/release layer 48 and allowed to bend up and away from
substrate 40 due to the stress gradient in metal layers 54a-n.
[0058] Another wet etchant which may be used is a buffered oxide
etchant (BOE) which is hydrofluoric acid combined with ammonium
fluoride. Also proper choice of the passivation/release material
can allow a dry-etching technique for the release process.
[0059] Metal patterns 56, 58 comprised of the same stressed metal
design of patterns 54a-n, are not released, and are used as sensor
readout lines and contact elements to the first
transparent/conductive layer 42 and second transparent/conductive
layer 46, which act as electrodes for sensor 53.
[0060] FIGS. 5e and 6e depict a sensor/contact semiconductor
integrated device 62 which carries sensor 53, an amorphous silicon
active device, together with stressed spring contacts 54a-n
designed to contact devices on a separate substrate.
[0061] It is again worth noting that substrate 40, first
transparent/conductive layer 42, second transparent/conductive
layer 46, and passivation/release layer 48 are each transparent at
the frequency of operation of VCSEL lasers 12 of printbar 10.
Sensor element 44 is partially transparent.
[0062] Interference is a phenomenon that can alter the reflection
from a surface. It can be designed beneficially to obtain
anti-reflection characteristics, reducing reflection losses.
[0063] Light being directed to sensor 53 may either be absorbed,
passed through, or reflected. Reflection of light is undesirable as
compared to the other possibilities, since if light is absorbed,
the sensor is using it to determine an appropriate feedback to the
system, and if light passes through, it is being used by the target
device, for instance to create a latent image on an electrostatic
drum or for other useful purposes. On the other hand, reflected
light is wasted light.
[0064] The interference phenomenon is dependent upon the thickness
of layers comprising sensor 53 and the wavelength used by laser 12.
Sensor 53 has been designed in consideration of the interference
phenomenon, and the thickness of the layers have been adjusted to
avoid or minimize undesirable reflection for the light frequency of
lasers 12. In particular, passivation/release layer 48 has a
thickness of 3,000 angstroms to obtain the desired non-reflective
effect.
[0065] Turning to FIGS. 7a-b and 8a-b, a second embodiment of the
present invention is illustrated. It is known that prior to a laser
starting its lasing process a phenomenon takes place known as
spontaneous emissions. During the spontaneous emissions, light in
the visible range from the laser may be emitted. It is undesirable
to have this light, as well as light of any other undesired
wavelength, reaching sensor element 44. Therefore, to further
improve the reliability of the present invention, when IR lasers
are used, an additional processing step may be added. Particularly,
after the application of second transparent/conductive layer 46 (as
depicted in FIG. 5b), a visible light absorption layer 64, which
may be hydrogenated amorphous silicon (a-Si:H), is deposited on top
of second transparent/conductive layer 46 prior to sensor element
44 and second transparent/conductive layer 46 being patterned.
Visible light absorption layer 64 is opaque to visible light, and
transparent to IR light. Once sensor element 44, second
transparent/conductive layer 46 and visible light absorption layer
64 have been deposited on top of first transparent/conductive layer
42, they are patterned. Next, and similar to FIG. 5c,
passivation/release layer 48 is deposited over this patterned
stack, and over transparent/conductive layer 42 and substrate 40.
Thereafter, and as shown in FIGS. 7a and 8a, vias 65 and 66 are
provided through passivation/release layer 48 and visible light
absorption layer 64, to provide access to transparent/conductive
layers 46 and 42. By this design, an electrically isolated sensor
67 is formed.
[0066] As depicted in FIGS. 7b and 8b, stressed metal layers for
patterns 54a-n and 56 and 58 are deposited in a manner similar to
that discussed in relationship to FIGS. 5d and 6d. Thereafter,
selected portions of patterns 54a-n are released by the etching
process previously discussed, to form integrated device 68.
[0067] The embodiment shown in FIGS. 7a-7b and 8a-8b adds visible
light absorption layer 64, which provides a manner of inhibiting
spontaneous emissions generated visible light from impinging upon
sensor element 44. This avoids false readings from sensor 67, which
would negatively impact the calibration process.
[0068] When the laser goes above the laser threshold, spontaneous
emissions may still exist, too. An ideal sensor would be "blind" to
the spontaneous emission component, i.e. it would have a very
narrow bandwidth. Therefore it would read nothing but the lasing
component of the laser operation. Absorption layer 64 is able to
absorb the continuing spontaneous emissions, so that it does not
reach sensor element 44.
[0069] In one embodiment, absorption layer 64 is a-Si:H of 1 micron
thickness, or preferably approximately 3,000-5,000 angstroms,
thick. Other materials having the capability of absorbing undesired
light and allowing the desired wavelength to pass may also be used.
It is noted that amorphous silicon will change in sensitivity
dependent upon the wavelength of light. By absorbing the visible
light, a more accurate reading is obtained. In a further
embodiment, visible light-absorbing layer 64 may be constructed
directly on the output of lasers 12.
[0070] Turning to FIG. 9, the cross section of an integrated device
70 is illustrated having a transistor, e.g. Thin-Film Transistor
(TFT) switch 72 configured below a semi-continuous sensor 74, which
is integrated with contact springs 76a-n, similar to spring
contacts 54a-n of FIGS. 5e and 6e. In this embodiment,
p-i-n-amorphous silicon (a-Si:H) sensor 53 of FIGS. 5e and 6e is
replaced by a more elaborate composition. The combination of TFT
switch 72 and semi-continuous sensor 74 are meant to be shown as a
pixel, or picture element of a 1- or 2-dimensional array, enclosed
in a layer of passivation, for operation as an active matrix
sensor.
[0071] With more particular attention to the construction of device
70, deposited on a transparent substrate 80 such as glass, is a
gate contact 82 formed of a transparent metal, such as Chronium
(Cr). Metal layer 82 is deposited in a thickness of approximately
3,000 angstroms, and acts as the gate contact of TFT switch 72.
Deposited over metal portion 82, and remaining portions of
substrate 80, is a first transparent/conductive layer 84, such as
nitride, oxynitride, polyamide or other appropriate material, which
is typically deposited to approximately 3,000 angstroms in
thickness. Deposited over layer 84 is a layer 86 of an intrinsic
hydrogenated amorphous silicon (a-Si:H), typically 500 angstroms
thick.
[0072] An island of nitride (oxynitride, polyamide, etc.) 88 is
deposited and patterned over gate contact 82 on the a-Si:H layer
86. Island 88 is typically deposited to a thickness of
approximately 2,000 angstroms.
[0073] A layer of n-doped a-Si:H 90 is then deposited and
selectively patterned to a thickness of approximately 1,000
angstroms over nitride island 88 and layer 86.
[0074] Next, a layer of transparent conductor 92 is deposited on
top of island 88, and an opening 94 is patterned to create two
electrodes 92a, 92b from layer 92. The metal of layer 92 may
typically be a Indium Tin Oxide (ITO) . Patterns 92a and 92b act as
the source and drain contacts for TFT transistor 72. A passivation
layer 96 is patterned on top of conductor layer 92 and may
typically by oxynitride of approximately 1 micron, or alternatively
a polyamide layer of approximately 2.3 microns thickness. A via in
layer 96 is opened, such that a transparent/conductive layer 98,
typically made of ITO, and an n.sup.+-doped amorphous silicon layer
100, are deposited and patterned in a mushroom-shape inside and
over the via. Layer 98 functions as the bottom electrode of sensor
74. Layer 98 is deposited such that, in the via, it is in contact
with layer 92 and over remaining portions of layer 96. The
n.sup.+-doped contact layer 100 is typically 700 angstroms
thick..
[0075] A continuous layer of intrinsic amorphous silicon (a-Si:H)
102 is deposited over the n.sup.+-doped contact 98 and portions of
the passivation layer 96. This layer of sensor 74 has a typical
thickness of approximately 1 micron.
[0076] A p.sup.+-doped layer 104 is then deposited over intrinsic
a-Si layer 102 to a thickness of approximately 100 angstroms. A
transparent/conductive layer 106, typically made of ITO and 5,500
angstroms thick, acts as a top electrode of sensor 74. Thereafter,
a top passivation/release layer 108 is deposited and patterned in
accordance with the description of FIGS. 5d-5e and 6d-6e, and metal
layers are deposited in accordance with the previous
embodiments.
[0077] This embodiment of integrated detector/contact spring device
70, therefore, consists of a TFT transistor 72 which is connected
to the semi-continuous sensor 74 through the opening created in
passivation layer 96. The sensor 74 is otherwise separated from the
TFT on a top level by portions of passivation layer 96 that have
not been etched away.
[0078] After formation of an integrated detector/contact spring
device, such as devices 62, 68 or 70, alignment is made with a
substrate carrying printbar 10 or the light source of interest. It
is then desirable to determine the performance of a system
configured by electrically contacting printbar 10 with driver chips
24, 26 (FIG. 3) through use of integrated sensor/contact spring
devices 62, 68 or 70. Therefore testing was undertaken to determine
the operating capability of sensor 53 in a system as described.
[0079] Initially, a given power was applied to a single laser 12,
and the output signal generated by sensor 53 was monitored as a
result of the input power. Typically, for 1 milliwatt of light
output, the signal of the photo-current provided by sensor 53 was
approximately 1 micro-amp of photo-current. The dark current, the
current that is produced when no light exists, was 1 pico-amp or
less.
[0080] For the sensor size suggested in the case of the 1,200 dpi-
the contrast ratio between the sensor current under laser
illumination and in the dark is therefore about 1,000,000, allowing
in principle for 10-bit resolution. In this embodiment, a 4-bit
correction has been used and can already provide substantial
quality improvement to the system.
[0081] Turning to the calibration process, it is noted that in a
first embodiment, calibration of lasers 12 of printbar 10 is
accomplished by sensing and calibrating a single laser at a time.
Particularly, sensors (34, 34', 53, 68, 74) are sufficiently sized
to be placed in front of all lasers 12 of printbar 10. In one
calibration scheme, the imaging device is not being used to print
an image during calibration. Rather, the calibration process takes
place during a time when image processing is not occurring.
[0082] In the embodiment describing the laser printbar, it is
assumed sensors 53, 68, 70 are rectangular sensors of approximately
4 cm by 200 micrometers, which is large enough to intercept 100% of
the laser beams diverging from printbar 10, for a substantially 4
cm-long laser array. The typical divergence of the VCSEL's beam was
noted to be smaller than 20.degree..
[0083] The transparency of the amorphous silicon film ensures
sufficient laser radiation to exit from the sensor to allow for
printing while low (10 pA/cm.sup.2) dark leakage current of sensors
53, 68, 70 maintains the contrast ratio (or light-to-dark ratio) at
a high enough value for operation.
[0084] Turning attention to FIGS. 10 and 11, a configuration
similar to that shown in FIGS. 3 and 4 is depicted, and the same
elements are provided with the same numbers. In this embodiment,
the printbar is an LED printbar 10'. It is noted that light from an
LED is non-coherent unlike the light from a laser and it is
typically much more diverging. This is illustrated by light beam
28' of FIG. 10. While an LED lightbar may be incorporated into a
configuration such as shown in FIGS. 3 and 4, the particular
characteristics of LED light are more fully taken advantage of in
the present embodiment.
[0085] Sensor 34', as shown more particularly in the bottom view of
FIG. 11, is manufactured having an elongated open "o"-shaped
configuration, one end of which is shown in the figure. It runs
along the full length of the LED printbar, with two leg portions
34a' and 34b', whereby an opening 34c' is provided over the LEDs
12' of LED printbar 10'. Such a design recognizes the diverging
characteristics of an LED light source, and positions sensor 34'
such that sensing elements 34a', 34b' are at the edges of the LED
light beam 28'. By this design, the direct LED light path, which is
emitted through the open section 34c', is undisturbed and therefore
substantially 100% of this light source may pass to the intended
target. Sensor 34', is able to detect the value of the light being
emitted by sensing the light in the shoulder or edge portions of
light beam 28'.
[0086] It is to be noted that using such a design, all of the light
from the shoulder portions of beam 28' may be absorbed and
sufficient LED light may still be emitted through opening 34c' to
allow sufficient light emission for the intended target. Depending
on the system it will be embedded in, this can allow sensor
configurations where the transparency to the light is almost zero.
In situations where the entire portion of the shoulder beams of
light 28' is not absorbed, due to the incoherent nature of the LED
light, that portion of the light passing through sensor 34' will
join the central portion of beam 28' which has not been disturbed
by sensor 34'.
[0087] The foregoing design is particularly useful in connection
with LED printbars since they have a less powerful light beam than
a laser light beam. Thus, by not absorbing the center part of the
light emission, a more efficient imaging system is provided.
[0088] Turning to FIGS. 12 and 13, an embodiment similar to that
shown in FIGS. 10 and 11 is provided. However, in these figures, in
addition to sensor 34' being provided with an opening, i.e. the
elongated open "o"-shaped configuration, a similar concept is
implemented with the substrate 32'. This design is particularly
illustrated in FIG. 13 where it is shown that substrate 32' is
provided with a glass opening 32a'. It is noted that while FIG. 12
appears to have substrate 32' as two separate elements, in
actuality and as illustrated more completely in FIG. 13 there is
simply an intermediate rectangular section of glass substrate 32'
which has been removed by any suitable etching technique, such as
wet etching. FIG. 12 describes a section across the integrated
device.
[0089] Turning attention to FIG. 14, a block diagram of a
calibration/printing system 110, according to the present invention
is depicted. Driver chip 24 (which could also be driver chip 26) is
shown in association with printbar 10 and sensor 34, (sensors 34',
53, 68, 70 or other appropriately formed sensor may also be
used).
[0090] The only data link between printbar 10 and sensor 30,
through which information is passed, is the light transfer. Sensor
34 generates a readout current I.sub.s which is carried on feedback
circuit lines 38, 39. This sensor readout current I.sub.s is
delivered to a comparator 112, and compared to an external
reference current I.sub.R. The value of external reference I.sub.R
is a parameter of the system set by a user or during system design.
Comparator 112 measures the difference between readout current
I.sub.s and reference current I.sub.R to obtain an offset current
I.sub.OFFSET, which is delivered to current/voltage converter 114,
and this voltage is in turn provided to analog-to-digital (A-/D)
converter 116. In this embodiment A-/D converter 114 is shown as a
four-bit A-/D converter. These four bits are routed to a set of low
frequency shift-registers being used as an electronic look-up table
118, which in this embodiment is comprised of four shift registers
118a-d. Each bit of data enters the shift registers and ripples
through as shown by arrow 120.
[0091] In one embodiment, for a printbar having approximately
14,000 lasers, each shift register 118a-d may be a 14 k-bit shift
having a serial input and a serial output. The outputs of shift
registers 118a-d are supplied, in parallel, as a 4-bit word, to
driver 122. In the example in FIG. 14, the output lines from the
top stage of shift registers 118a-d are delivered to the driver at
a stage 1 position 124, via input lines 125a-d.
[0092] In an embodiment with 14,000 lasers, there will be
approximately 14,000 stages each associated with a specific laser
of printbar 10. Therefore, each stage is connected or associated
with a specific laser. For each stage, e.g. stage 124, four bits
from (MSB to LSB) are provided by shift registers 118a-d. Each
4-bit value in each stage acts as the correction value for that
particular laser. It is to be appreciated that while four bits are
described in this embodiment, systems with a larger or smaller bit
number may also be implemented.
[0093] In the above-described section, the steps from sensing data
representative of laser output, until a correction value is loaded
into one of the stages of driver 122, are accomplished at a
comparatively slow rate. For example, a calibration operation as
described for all lasers in a 14,000 array may take approximately
1-2 seconds.
[0094] In the case of a printbar, since calibration can be
programmed to take place upon either start-up of the machine,
during a rest period, or at predetermined times when the machine is
not operating, the time to acquire and store the information into
4-bit driver 122 is not critical. Through this process 4-bit driver
122 has its inputs set to the correction values for each laser of
printbar 10. It is to be appreciated, however, that while this
embodiment uses a separate time sensor readout and correction of
the driver's inputs, with sufficiently timed actions and
appropriate data handling, high-speed real-time, or near real-time
calibration may be accomplished with a similar approach.
[0095] High-speed printers run data streams at a frequency much
greater than that just described for this embodiment. Even if the
calibration operation is not designed to keep up with the speed of
the high-frequency data stream, by the process now described, this
differential in speed is not critical as the correction values
stored in the stages such as stage 124, are already set at the
inputs of the driver 122 when the high frequency data stream from
shift register 126 is enabled. The values are set to stage 124, via
input lines 125a-d from low frequency shift register 118.
[0096] Turning now to a printing process using a printbar with
approximately 14,000 lasers, data will be supplied via a
high-frequency bit stream 127, which may be provided through a
print processor of a computer or other digital device. The high
frequency data is supplied to a high frequency 14 k-bit shift
register 126 with a parallel output to form enable/disable outputs
126a-n. The correction information stored at the inputs of stage
124 is used to adjust, with respect to a predetermined mean value,
an activation signal 130a-n supplied to printbar 10, in order to
generate an appropriate level of output for the corresponding
laser. The outputs 126a-n of high frequency shift register 126 are
used to enable/disable the corrected current, for each stage, from
being delivered from the driver to the corresponding laser. It is
noted that in this embodiment 126a-n represents approximately
14,000 outputs and 130a-n represents approximately 14,000
signals.
[0097] The correction value set at the inputs of the stages, such
as stage 124, therefore are stable values, held in the electronic
look-up table 118.
[0098] Returning attention to the calibration procedure, lasers 12
are, in this scenario, activated in a sequential one-at-a-time
fashion and sensor 34 reads out the photo-current produced by each
laser one at a time. The time response of the photodiode to the
radiation pulse is virtually immediate and is limited by the
readout electronics. The photo-current turn-off is related to the
transient time of holes (slower photo-carriers) through the
depleted intrinsic region. As illustrated in FIG. 15, amorphous
silicon transport properties very safely allow readout times of
about 10 .mu.s, 132. A 100 .mu.s idle time, 134 for a VCSEL type
laser combined with an amorphous silicon p.sup.+-i-n.sup.+
photo-diode is considered desirable for clean operation. Given a
sensor capacitance of 0.8 nF (as calculated from a 10 nF/cm.sup.2
typical capacitance), the resistance of a read-out circuit can be
as large as 1 KOhms to widely maintain the 10 .mu.s readout
constraint. In reality, the current is flowing into a virtual
ground and resistance can therefore be made small.
[0099] The use of a single sensor for all the VCSEL lasers allow
for an ease of fabrication and correct normalization of the power
outputs even if the dark sensor leakage current would tend to
degrade as the sensor ages. One reason why the sensor may degrade
is illumination-induced defect creation (known as Staebler-Wronski
effect). Defect creation will also alter the photo-response of the
sensor. Since the same amount of radiation output (in a duty cycle)
is expected on average for all the lasers during printing, this
effect should not affect the system's performance. The use of a
continuous sensor layer increases interaction between areas
illuminated by adjacent VCSEL lasers and averages the previously
discussed effect.
[0100] To prevent Staebler-Wronski effect from disturbing a correct
calibration operation also, the sensor can be pre-degraded before
use by exposure to light for an appropriately long time. Typically
Staebler-Wronski effect degrades the sensor photo-response to about
70% of its initial value. Thereafter, the effect does not cause any
further degradation and the performance of the sensor is to be
considered permanently stable. The small reduction in the response
to illumination is completely irrelevant given the extremely large
on-to-off ratio previously noted. The thickness of a amorphous
silicon (a-Si:H) sensor is to be uniform to 2-3% on the area of
interest. The effect in the uniformity on the intensity of the
transmitted radiation is therefore small. Further, it constitutes a
pattern at low spatial frequency, not particularly relevant for the
human eye.
[0101] For a 14,000 VCSEL array, the total calibration time is
approximately 1.54s, (i.e. 14,000.times.110 .mu.s) on the basis of
a design according to the first embodiment of a single sensor.
[0102] To evaluate the effect of this time on throughput, these
statistics are considered with a 600-page/minute printer. For such
a printer, a page time of 100 ms exists. Therefore, re-calibrating
every 1000 pages (i.e. slightly less than once a minute) on such a
high-speed printer would reduce the pages per minute output to 591
which is a 1.5% reduction in system output. The calibration
per-1000-pages is not considered a requirement to detect laser
aging and other problems. Therefore, considerably slower
calibrations may be safely adopted and throughput will not be
significantly affected even at these very high-speed printers.
[0103] The actual control of printbar 10 during printing operation
is a more critical step due to a higher speed of signals in
high-speed printing (e.g. 673 Mhz when two drivers are employed).
An embodiment of a driver element 140 of driver 122 for one laser
12, which may be used for this purpose, is shown in FIG. 16. Driver
140 is based on CMOS electronics and has been designed for 4-bit
uniformity correction. Such a device consists of a set of
appropriate current mirrors, properly gated, and of the actual
VCSEL driver stage. Each current mirror is comprised of a set of
transistors 142a-e, and a set of reference transistors 143 which
are shared by all sets 142a-e. Each current mirror 142a-d is
programmable by means of a set of the gating input lines 125a-d
Gate input lines 125a-d are driven by the outputs of low frequency
shift registers 118a-d. The chosen current correction adds to the
reference current delivered by the current mirror 142e. The total
output current to the light source (laser, LED or other similar
one) is delivered by activation signal 130 by the output stage of
the driver 122. Output is enabled by enable signal line 126a, of
enable signal lines 126a-n of FIG. 14, when appropriate.
High-frequency data are fed into high-speed shift register 128 and
enabled out in parallel to print an entire line of an image. With
continued attention to FIG. 16, activation signal 130 corresponds
to activation signal 130 of FIG. 10.
[0104] In a further embodiment of the invention described above,
background correction to suppress the small effect of the sensors'
leakage current may be provided. This can be relevant when a
many-bit correction is desired, such as 8 or 10-bit correction.
[0105] Another improvement can be realized by choosing an
appropriate firing order for the VCSELs (or any other similar light
source) during calibration, as opposed to a linear scan from one
end to the opposite one. This assists in reducing the effect of
local trapped charge in the intrinsic layer, relatively slow to be
released and that might distort the local electric field and alter
the photo-current. For example, this may be accomplished by firing
alternatively opposite ends of the printbar. An appropriate storage
path into the look-up table must be adopted accordingly.
[0106] The calibration process described is closed-loop but allows
for only one cycle correction. In order to further optimize the
system, more feedback cycles for each VCSEL may be added. This can
be done either by repeating the process more than once and
eventually adjusting the look-up table content further, or
extending the duration of the VCSEL test pulse to more than one
sensor readout time to obtain the same goal. The latter option
(faster) requires changing the set of four shift registers to a
slightly more elaborate system of individually addressable latches
to repeatedly adjust the content of any cell among the 14k. Both
solutions would slow the calibration process, although still in an
acceptable way in order to keep a high throughput in printing.
[0107] An alternative to the uniformity correction obtained by
drive current adjustment relies on the typical low duty cycle of
the VCSEL pulse. A proper modulation of the duty cycle, as dictated
by the look-up table values, will in fact provide a time-domain
uniformity correction without altering the laser drive current.
[0108] In a further embodiment, sensors 34, 34', 53, 68, 70 may be
constructed as a plurality of sensors, into a sensor array. In this
manner, instead of testing a single laser at a time, multiple
lasers of multiple arrays may be tested in parallel. A drawback of
using smaller sized arrays as opposed to a single sensor is that
the sensor medium may age at different rates for different arrays
used. An advantage is that the speed of the calibration process is
increased by parallel operation and makes easier to push the
calibration procedure toward real-time.
[0109] It is to be appreciated that the many aspects of the
discussion concerning driving a VCSELs printbar also apply to LED
printbars and to single light sources or arrays of light sources
that can be contacted by micro-springs and monitored by a proper
sensor configuration integrated with the micro-springs.
[0110] A micro-spring interconnect structure consists of
micro-springs used to electrically connect two or more devices.
Such a structure can be used in embodiments where the springs are
anchored on any of the devices being contacted. Therefore it is
also to be appreciated that the micro-springs, for example in FIG.
3, can be fabricated on the GaAs chip 10 instead of on the glass
substrate 32.
[0111] The foregoing is considered as illustrative only of the
principles of the invention. Further, since numerous modifications
and changes will readily occur to those skilled in the art, it is
not desired to limit the invention to the exact construction and
operation as shown and described, and accordingly, all suitable
modifications and equivalents may be resorted to falling within the
scope of the invention.
* * * * *